Hey, I'm Bayu 👋
I'm a Senior Backend Engineer based in Indonesia, building AI platform infrastructure that actually works in production — not just in demos. 10+ years in the trenches, currently focused on cost-aware, multi-tenant systems that scale to zero when idle.
What I Do
I design systems that survive production constraints. Unpredictable traffic, strict authorization, budget limits — that's where I thrive. Most of my recent work revolves around LLM infrastructure: multi-tenant gateways with per-service cost tracking, shared data source authorization via SSO/IAM, and observability stacks that give you real answers instead of dashboards full of noise.
My approach is opinionated but pragmatic: favor simple scaling models over clever ones, isolate failure domains early, and treat cost as a first-class architectural constraint — not a finance problem discovered later.
Background
Started as a full-stack developer, went deep into backend systems and infrastructure, and now I'm building the next layer: AI platforms that enterprises can actually trust. I've led architecture and implementation of AI platforms serving enterprise clients with strict security and compliance requirements — from design reviews and tradeoff decisions to production incidents and long-term evolution.
Outside of work, I run Nodestral — a fleet management agent I built to monitor my own VPS infrastructure. It started as "I don't want to SSH into 5 servers" and turned into a Go binary that discovers services, containers, and system metrics across multi-cloud deployments.
Tech I Work With
Languages
- Go (primary)
- Python
- TypeScript
- Kotlin
Infrastructure
- Kubernetes (AKS, EKS)
- Docker / Podman
- Terraform
- OpenTelemetry
AI / LLM
- LLM Gateway Design
- Multi-tenant Routing
- Cost Tracking & Observability
- Provider Integration
Backend
- PostgreSQL / Supabase
- Redis
- Clean Architecture
- Event-driven Systems
Projects
- Nodestral — VPS fleet management agent in Go. Discovers services, containers, and system metrics across multi-cloud deployments. Single 10MB binary.
- LLM Gateway — Production-grade multi-tenant LLM gateway in Go with cost tracking, rate limiting, and OTel observability.
- AI Observability — Grafana + OpenTelemetry monitoring suite for LLM workloads. Latency, token usage, and cost visibility.
- Interview Simulator — AI-powered interview practice. Upload a JD → generate tailored scripts → practice with real-time feedback.
What I'm Looking For
A remote Senior AI Platform Engineer role at a company building developer tools or AI infrastructure. I want to own systems end-to-end: architecture, implementation, production operations, and evolution. Bonus points if the team cares about cost efficiency as much as performance.
"I design cost-aware AI infrastructure that scales to zero, with enterprise-grade security and multi-tenant isolation."